The ability to prepare and manipulate organic thin-film-metal structures in a well-defined way is critically important in areas as diverse as sensing, adhesion, tribology, corrosion and nanoscale electronics. For example, FTIR spectroscopy in the Attenuated Total Reflection mode (ATR-FTIR) is a widely used surface analytical method for investigating an organic-metal interface. One of the most popular ATR attachments employs a 45xc2x0 ZnSe single crystal with an available sample area of 10 mmxc3x9770 mm (Horizontal Contact Sampler, Spectra-Tech, Inc.). In order to obtain the maximum possible ATR-FTIR signal strength, it is necessary to ensure maximum interfacial contact between the ZnSe crystal and the sample. Thin, smooth metallic foils are, therefore, a practical requirement for such surface studies.
Many types of thin metallic foils necessary as substrates in this area of research are commercially available but the cost of such materials is frequently too high for researchers with limited resources. For instance, single crystal pieces of any metal, whether precious or non-precious, and polycrystalline pieces of all precious metals typically cost at least about $1500 cmxe2x88x922. Thus, each 4 cm2 piece of commercial thin copper foil costs about $16 and many such pieces must be used during an investigation. Accordingly, an inexpensive way to obtain high quality thin metallic foils is desirable.
There has been a surge of interest in fundamental studies involving the properties in chemical reactivity of organic films on metallic copper evident in the recent scientific literature. A method of producing thin copper foils is therefore expected to be of particular interest to those presently involved in organic-copper interfacial chemistry, as well as people who are contemplating entry into this area.
This invention provides a new method for the preparation of thin metallic foils and organic thin film-metal composites using such foils. More particularly, the invention provides a method in which a metallic substrate is rotated and contacted with an etchant so as to reduce the thickness of the metal substrate by corrosion and realize the desired foil thickness. The resulting thin foil can then be contacted with the organic material so as to form a thin film of the organic material on the surface of the foil.
The process of the present invention is applicable to any metal substrate. While it can be applied to precious and rare metals, because the process depends on a controlled corrosion, it is generally impractical for such metals because of the cost involved. A non-precious metal is therefore preferred. At present, the most preferred metal to which the process of the present invention can be applied is copper.
The thickness and other dimensions of the metal substrate employed are not restricted as long as the substrate can be rotated. Relatively thick metallic foils are commercially available and the use of such substrates is generally preferred. In general, it is preferred to employ a substrate having a thickness in the range from about 0.01 to about 1 mm and preferably from about 0.1 to 0.6 mm. It is also generally preferred, although not essential, that the substrate employed be in the form of a circular disk.
The metal substrate is rotated in any convenient manner, for instance by being mounted on a suitable rotatable supporting structure. The mounting also serves to protect one of the major faces of the substrate from the etchant and thereby minimize point breakthroughs which are more likely to occur when two opposing surfaces are being simultaneously etched. The speed of rotation can be adjusted as desired depending on the particular metal substrate employed, the particular etchant, the temperature and pressure conditions employed, and the desired rate of etching. In general, the rotation is usually within the range of about 100-1000 rpm, preferably 200-600 rpm.
Any etchant which acts chemically on the metal substrate to remove surface from that substrate can be employed. In general, inorganic acids such as hydrochloric acid, sulphuric acid, nitric acid and hydrofluoric acid can be used, with nitric acid being particularly preferred, but the range of etchants usable is not limited to this list. The etchant and metal substrate are contacted under etching conditions. For some combinations of etchant and metal, elevated temperature and/or pressure may be appropriate while other combinations can be processed under ambient or other conditions.
The etchant and metal substrate are contacted under the etching conditions for a time sufficient to reduce the substrate thickness to the foil thickness desired. If desired or convenient, the contacting may be done a plurality of times until the desired foil thickness is achieved. For example, the metal substrate can be separated from the etchant periodically and the exposed surface washed or otherwise cleaned before contact with the etchant is reestablished. The thickness of the desired thin foil is typically on the order of about 1 to 100 xcexcm and preferably about 10 to 90 xcexcm. Frequently about 60-90% and more frequently about 70-85% of the initial mass of the metal is eliminated as a result of the etching procedure.
Forming an organic thin film on a thin metal foil is per se known. The foil and the organic material are brought into contact under suitable conditions so as to form an organic thin film on the surface of the metal foil. Any procedure and any organic material which has been used heretofore can be used in accordance with the present invention which differs from such prior procedures by using the thin metal foil produced by the rotating etching procedure described above.
In order to further illustrate the present invention, experimental procedures and results are described below. It will be understood that these examples are intended to be illustrative and non-limiting. Unless otherwise indicated, all temperatures are in degrees Centigrade and all parts and percentages are by weight throughout this specification and claims.
General. Kinetic data were obtained with copper sheet (Johnson Matthey, 0.60 mm thickness, 99.9% purity). Copper xe2x80x9cHeavy Foilxe2x80x9d (J. T. Baker, 0.127 mm thickness, xe2x89xa799.90% purity) was used for all other experiments. Water was deionized (DI) via reverse osmosis (Barnstable Nanopure II system, resistivity 16.7 Mxcexa9). Anhydrous ethanol (Phannco), hexane (Fisher, HPLC grade), acetone (Fisher, reagent grade), dichloromethane (Fisher, HPLC grade), dodecane (Aldrich, 99.9+%), concentrated nitric acid (Fisher, reagent grade), dodecanethiol (Aldrich, 98+%), 11-mercapto-1-undecanol (Aldrich, 97%) and acetyl chloride (Aldrich, 97%) were used as received. Visible spectroscopy was carried out on a Milton Roy Model Spectronic 20 spectrophotometer. FTIR spectra were obtained on a Mattson Galaxy Model 3000 spectro-photometer, using Win 1st software for processing (DTGS detector, 2-cmxe2x88x921 resolution). ATR-FTIR spectra were obtained using a Horizontal Contact Sampler (Spectra-Tech, Inc.) fitted with a 10-mmxc3x9770-mm ZnSe crystal (45xc2x0, 12 internal reflections). Two 10 mmxc3x9735 mm strips were cut from each sample and were laid end-to-end on the ZnSe crystal before application of pressure by the gripper. The ATR-FTIR sample compartment was purged with nitrogen before analysis of samples. IR spectral assignments were assigned, in part, with the aid of molecular modeling data (PC Spartan Plus(trademark), Wavefunction, Inc.), using a PM3 basis set.
Kinetics. A circular disk 48 mm in diameter was cut from copper sheet of nominal 0.60-mm thickness (8.28 g, 0.130 g-atom, 99.9% purity). The center of the disk was fastened to the circular end of a wooden dowel (6 mm in diameter, 15 cm in length) with two-part xe2x80x9c5-minutexe2x80x9d epoxy adhesive. After the adhesive cured, the disk/dowel assembly was attached to a digital stirring motor (IKA Tech, Inc., Eurostar(trademark) model). The disk was vigorously abraded with emery paper and then lowered into 6 N HNO3 (75 mL, 0.45 mol, contained in a 4-mmxc3x9780-mm crystallizing dish). The disk was allowed to rotate at room temperature for 5 minutes at a total spinning rate of 500 rpm. Evolution of hydrogen gas was evident, as was the development of the characteristic blue color of aqueous Cu(NO3)2. The resulting solution showed a maximum in the visible spectrum at xcex=370 nm. The copper disk was rinsed with DI water, then lowered into 75 mL of fresh 6 N HNO3. The rotating rate was adjusted to 250 rpm, and 3.1-mL aliquots were removed at periodic intervals for spectroscopic analysis at xcex=370 nm. Absorbance values for each kinetic point were corrected by a factor of (solution volume at time t)/(initial volume of solution) to compensate for the volume of each aliquot removed. The observed pseudo-first order rate constant, Kobsd, was determined from the semilog plot of the corrected absorbance values, Abscorr, vs. time, t.
Standard Etching Protocol. Hydrogen gas and NOx fumes are rapidly generated in the initial phase of this exothermic reaction. Therefore, this procedure was carried out in a good fume hood, using appropriate personal protective gear, and ensuring that no potential sources of ignition are present. A circular disk was cut from the copper xe2x80x9cHeavy Foilxe2x80x9d of nominal 0.127-mm thickness (xe2x89xa799.90% purity). The disk diameter was 60.0 mm (3.45 g Cu). The disk was fastened to the end of a wooden dowel as described above. The disk/dowel assembly was attached to the digital stirring motor, and the disk was carefully lowered onto the surface of a cooled nitric acid solution (30 mL, 7.5 N, ice water bath, T≈20xc2x0 C.) contained in a 80 mmxc3x9740 mm crystallizing dish. Stirring was immediately begun (200 rpm). The corrosion reaction of copper was quite vigorous initially. Hydrogen gas and brown NOx fumes evolved from the metal-solution interface; the first-formed product had a deep emerald-green color before its complete dispersal into the bulk medium. An additional 15-mL portion of DI water was added within ca. 15 s, and the disk was raised so that the lower disk surface continued its rotation at the liquid-air interface. At that point, the volume of solution was ca. 45mL and the HNO3 concentration was 5.0 N. The solution developed an intense blue color as reaction progressed. Stirring was allowed to continue for an additional 7-8 min; the size of the evolving H2 gas bubbles appeared constant (≈0.5 mm) throughout the reaction period. Stirring was discontinued and the thin copper disk was slowly raised from the liquid. After quickly rinsing with DI water and acetone, the disk was allowed to air dry. Using a scalpel, the thin-foil disk was carefully detached from the epoxy adhesive that had fastened it to the dowel. The foils were subsequently handled only with polypropylene tweezers. The thin foils were either stored in polyethylene bags, stored in the open atmosphere, or used immediately for chemical adsorption experiments.
Adsorption Experiments. Thin copper foils were subjected to a preadsorption protocol immediately prior to chemical adsorption regimens. This preadsorption protocol consisted of a quick treatment with 2 N HNO3 (ca. 5 mL for 2-10 s), followed by rinses by DI water and either anhydrous ethanol or acetone. The foils were then treated within ca. 5 s with chemical reagents in adsorption experiments (either as 0.05-020 M ethanolic solutions or as neat liquids). While not effected in an oxygen free system, high quality organic material monolayers could be obtained as long as the exposure to air was minimized between the etching, rinsing and absorption steps.
Kinetics. In the kinetics experiment, a plot of [In(Abscorr)] vs. time (t) gave a straight line, with an observed pseudo-first order rate constant of 0.038 sxe2x88x921 (txc2xd=18 s). This pseudo-first order kinetic behavior is expected for a diffusion-controlled reaction at a rotating disk. Immediately after the etch reaction, the copper surface had a glittery, pinkish appearance. After rinses with DI water and acetone, tarnishing of the dry surface became evident within ca. 25 min. Additionally, the copper disk displayed the characteristic swirled, pattern of etching. The final thickness of the disk was ca. 50 xcexcm. This particular disk was neither flat nor malleable; it was only used to demonstrate that the expected kinetic behavior was obtained.
Standard Etching Protocol. The Standard Etching Protocol described above successfully and reproducibly provided thin, flat, smooth foils of metallic copper of reasonably uniform thickness. The thickness of each foil thus prepared was measured at four points at the edge of the piece and at its center with a micrometer. Typical values were: 10 xcexcm, 15 xcexcm, 28 xcexcm, 12 xcexcm (four equally-spaced points around the perimeter); 82 xcexcm (at the center). It is noteworthy that 70-85% of the initial mass of copper was lost as a result of the controlled etch. The initial appearance of the resulting thin copper foil was smooth, glittery and rose pink in color. The individual grains of the polycrystalline metal were visible to the eye. Each grain appeared to be ca. 0.01-0.10 mm2 in area.
Adsorption Experiments. Treatment of the prepared thin copper foils with ethanolic solutions of dodecanethiol (DT) or 11-mercapto- 1-undecanol (MUD) afforded samples which were characterized by their corrosion and wetting behavior, and by their ATR-FTIR spectra. In the case of MUD on copper, subsequent chemical reactivity of the putative hydroxy-terminated monolayer was also tested.
Corrosion protection of metallic copper was conveyed by both DT and MUD, as judged by visual inspection of samples which were stored in polyethylene bags. DT on copper (DT/Cu) showed little evidence of tarnishing over a 12-month period. In contrast, some darkening of the MUD/Cu samples was apparent over a period of 1-2 weeks. These observations generally agree with results reported previously. DT/Cu samples that were gently swiped with a dry lab tissue showed a tendency to darken within hours at the location of mechanical disruption, indicating reoxidation of the metal.
DT/Cu is completely wetted by dodecane and not wetted by DI water (50 xcexcL, sessile drop). MUD/Cu is completely wetted by dodecane and partially wetted by DI water (contact angle≈130xc2x0, initially: the contact angle with water decreased to ca. 80xc2x0 after 22 hr). The general wetting/repellancy behavior of both DT/Cu and MUD/Cu is consistent with contact angle measurements previously reported.
The results of ATR-FTIR analyses of DT/Cu and MUD/Cu samples, as well as IR data for the absorbates, are presented in Table I and Table II. Generally, lower frequency skeletal stretches and bends were easily discernible for these samples (1500-600 cmxe2x88x921 region). The appearance of new absorbance bands in this region, characteristic for Cxe2x80x94C and Cxe2x80x94O stretching modes, CH2 bending modes, and related coupled vibrational modes, strongly suggests that organic thin-film/copper structures were formed. For each sample, however, xcexd(Sxe2x80x94H) was absent, which is consistent with the formation of a covalent Cu(I)-thiolate linkage on the metallic surface.
The experimental IR signals can be compared with computational predictions for the model mononuclear compounds dodecylthiolatotriphosphinecopper(I), (CH3xe2x80x94(CH2)11xe2x80x94Sxe2x80x94Cu(PH3)3, (A in Table III) and 11-hydroxyundecylthiolatotriphosphinecopper(I), (HOxe2x80x94CH2)11xe2x80x94Sxe2x80x94Cuxe2x80x94(PH3)3, (B in Table III). Because of the overall broadness of the experimental IR absorptions in the 1350-600 cmxe2x88x921 region and the low signal strengths involved (Table II), it is not possible to make definitive assignments. It can be concluded, however, that the experimental data generally agree with the computational predictions, further supporting the conclusion that thin copper foils prepared by the protocol described are viable substrates for chemisorption with DT and MUD.
Freshly prepared MUD/Cu assemblies undergo additional chemical derivatization. When treated with acetyl chloride (neat, followed by an ethanol rinse and air drying), the resulting organic-copper samples exhibit the expected changes in the ATR-FTIR spectrum, with characteristic signals appearing for the new ester moiety: CH3xe2x80x94C(O)xe2x80x94Oxe2x80x94(CH2)11xe2x80x94Sxe2x80x94Cus.
Clean, reactive metallic copper surfaces can b prepared in a well-defined manner by taking advantage of the hydrodynamics of he rotating disk system. Copper disks can be corroded uniformly to produce thin, reactive metallic foils, to which organic materials adhere. The nature of the resulting organic films on copper, as examined by ATR-FTIR spectroscopy and by wetting and corrosion-protection behavior, is the same as for those reported previously.
The standard etching protocol was developed it order to (1) product flat, chemically reactive metallic copper pieces; and (2) produce thin and flexible copper pieces so that optimal interfacial contact between a copper thin foil-organic sample and the ZnSe ATR crystal would result. These two practical requirements, which are sine qua non for obtaining useful ATR-FTIR data, have been achieved. The absorbance values obtained in spectra of DT/Cu and /Cu samples strongly imply the existence of chemisorbed monolayers. Thus, the chemical viability of copper surfaces which have been prepared according to the controlled etching and preadsorption protocols has also been demonstrated. The reactivity of MUD/Cu samples towards acetyl chloride confirms that addition 2-dimensional chemistry is accessible in suitably functionalized metal-organic assemblies.
It is important to note that all experimental work described herein has been carried out with readily available and relatively inexpensive materials, chemical reagents, analytical instruments, computer resources an molecular modeling software. Thus, meaningful surface science research can be carried out on a limited budget.